Determining Constraints of Moving Variance to Find Global Optimum and Make Automatic Clustering
Determining Constraints of Moving Variance to Find Global Optimum and Make Automatic Clustering
复制标题
确定移动方差约束寻找全局最优并进行自动聚类
DOI：
--
复制
发表时间：
2004
期刊：
影响因子：
--
通讯作者：
K. Arai
中科院分区：
文献类型：
--
作者：
Ali Ridho Barakbah;K. Arai
本文提出了一种寻找聚类全局最优的新方法。分析了聚类构建各阶段的移动方差，观察了聚类构建的全局最优模式，避免了局部最优。该算法引入了两种约束条件，即山谷跟踪和爬坡，以寻找全局最优解。此外，本文还分析了自动聚类的可能性。实验结果验证了本文方法的有效性。
This paper proposed a new approach to find the global optimum of clustering. It analyzes the moving variance of clusters for each stage of cluster construction, then observes the pattern to find the global optimum as well as avoid the local optima. It introduces two constraints, valley-tracing and hill- climbing, to find the global optimum. Besides this paper analyzes the possibility to make automatic clustering. Experiment result performs the effectiveness of the proposed approach in this paper.